Reduce the Azure Cloud Spend $$$ using Citrix Cloud Cost Optimization features.

Hi All,

In this blog we are going to see the different methods we can leverage Citrix Cloud to optimize the Azure Cloud Spend for VDI/XenApp environments.

Reduce the Azure disk charges using Storage cost savings option.

Citrix has released the “Storage cost savings” option for persistent VDIs which will essentially convert the VM disk from its original state to HDD when the machine is in powered off state. We can configure the “storage cost savings” option during the Machine catalog creation process. I have tested it with Single session MCS catalog. The option is not available when we select multi session OS.

To configure this option, navigate to Citrix Cloud –> DaaS–>machine catalogs and click on create new catalog. Select the catalog type as single session and navigate through the different options available in the creation wizard.

Tick the “Enable storage cost savings” option under disk settings to enable the storage cost savings option.

I created a standard SSD machine and the below screenshot is from Azure when the VM is in running state.

Now Let us see how this helps reduce the cost $$$.

Assume that you are running 100 persistent VDIs in your estate and all are running premium SSDs.

The cost of 100 premium SSD disks are $19.71*100 disks which equals to $1971,you need to pay the disk cost irrespective of your VM state.

Now let us assume that your average VDI usage hours are 12hr/day. So, you can shut down the VMs for the remaining 12hrs. The cost of 100 standard HDD disks are $5.89*100 disks which equals to $589.

So, if you can degrade your disks during the remaining 12 hrs. to HDD, the average disk cost you will be saving will be approximately $690. This is exactly how Nerdio is bringing value to AVD customers.

Note:_ the values are based on azure calculator without applying corporate discounts

Leverage Autoscaling feature to shut down the VMs during off-peak hours..

Autoscaling feature is supported in both multi session and single session VDAs.Auto scaling feature technically does the power management of the VMs to save the cost and energy.Auto scaling feature works in align to the remote desktop session timeout settings. We need to configure the idle session timer for auto scaling to function properly as we are providing actions via auto scale to shutdown or suspend the VM when it is in disconnected or logged off state.

To set the auto scaling feature, click on the delivery group and select “manage auto scale” feature from the top options.

In the general tab, enable the tick box in “enable auto scale” option. You can configure power off delay and also provide the average cost of VM which will be used by Citrix cloud to calculate your average savings.

In the “schedule and peak time” tab, select the peak and off peak hours during weekdays and weekends.

In the Load based settings tab, select the disconnected and log off action and also add the capacity buffer. capacity buffer normally will be used in multi session catalogs.

How this will reduce the cloud spend?

Let us assume that you are running 100 Azure Ds2v5 VDIs and not configure any power management options. So technically your VMs are running approx 730 hours a month. The cost of the VMs after a month will be $7008.

Let us assume that you have enabled power management to shutdown the VMs after the 14 hours of average user working hours.

So technically your VMs are running approximately 430 hours a month. The cost of the VMs after a month will be $4128.

With auto scaling feature you are saving an average of $2.8K per month. Now if we combine the storage cost savings option, then your cloud spend can be reduced significantly.

Hope this blog helps to optimize your Cloud Spend and get a better ROI.

Advertisement

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s